The Vaginal Infections and Prematurity (VIP) Study, jointly sponsored by NICHD and NIAID, was a prospective study designed to investigate the relationship between genital tract colonization with various microorganisms and the subsequent development of preterm birth. It also incorporates a clinical trial of erythromycin to prevent preterm birth among women colonized with Ureaplasma urealyticum, Chlamydia trachomatis, and group B streptococcus. In addition, the VIP collected a wealth of additional data on a variety of factors possibly linked to pregnancy outcome. Analyses undertaken to date include the predictive value of vaginal Gram stain in the identification of group B streptococcus, the descriptive epidemiology of group B streptococcal carriage, the relationship between reported physical activity and preterm birth, the effect of treatment with erythromycin on pregnancy outcome among women colonized with group B streptococci, and the association between sexual intercourse during pregnancy and preterm birth among women colonized with different genital microorganisms, effect of group B streptococcal colonization on pregnancy outcome, the association between bacterial vaginosis and pregnancy outcome, the effect of treatment with erythromycin on pregnancy outcome among women colonized with Chlamydia trachomatis, and ethnic differences in genital flora. During 2004, we investigated the association between ethnicity and vaginal pH (elevation of vaginal pH has been associated with adverse pregnancy outcome). We found that among women with normal vaginal flora, blacks had slighly elevated pH compared to whites. However, we also noted that even among the normal range, black women had higher vaginal Gram stain scores (indicating more abnormal flora) than their white counterparts. At comparable values of Gram stain score, the pH of black and white women did not differ significantly. This suggests that pH differences are due to differences in flora, rather than being a directly innate characteristic. We also evaluate whether change in vlora between 26 weeks gestation and delivery is associated with preterm birth. Acquisition of the Gram-negative rods E.coli and K.pneumoniae was associated with a roughly tripled risk of preterm birth. For no other organism was acquisition associated with preterm birth.